Recent advances in chemotherapy for pancreatic cancer: evidence from Japan and recommendations in guidelines

In Japan, along with the rapidly aging population, the number of patients with pancreatic cancer is also rapidly increasing, similar to the case for lung and colorectal cancer [1]. The number of deaths from pancreatic cancer in Japan has recently exceeded that from liver cancer, and pancreatic cancer now ranks as the fourth leading cause of death from cancer. In the majority of patients with pancreatic cancer, the cancer is already at an advanced unresectable stage at the time of diagnosis. Even in patients with resectable tumor at diagnosis who are treated by surgery, recurrence often occurs in the early phase after the operation. Thus, the prognosis of pancreatic cancer remains extremely poor [2].

The two most important approaches to improve the prognosis of pancreatic cancer are (1) to establish a better diagnostic method that would enable detection of this cancer at a resectable stage, and (2) to establish effective non-surgical treatments for patients with recurrent or unresectable pancreatic cancer. Until recently, there were a few effective non-surgical treatments. However, numerous clinical studies have been conducted in recent years, which have led to the establishment of new standard treatments that offer survival advantage. Several phase III studies have also been conducted in Japan, which have led to the establishment of standard treatments unique to pancreatic cancer patients in Japan.

In this article, we review the recent developments in chemotherapy for pancreatic cancer, including those based on evidence from clinical trials conducted in Japan, the current standard therapies recommended in guidelines in Japan (Figs. 1 and 2) and overseas, and global attempts to establish new strategies to overcome this disease with a grim prognosis.

Patients with pancreatic cancer are known to show high recurrence rates even after curative resection, and the prognosis of patients with recurrent disease is extremely poor. To prevent or delay the development of recurrence after resection and to improve the prognosis in patients with resectable tumor, numerous clinical trials of adjuvant therapy, including chemotherapy and chemoradiotherapy, administered before and/or after resection, have been actively undertaken both in Japan and overseas. Among the several types of adjuvant therapy, postoperative adjuvant chemotherapy has come to be recognized as a standard treatment strategy globally, based on demonstration in recent phase III studies of its ability to improve the long-term prognosis of pancreatic cancer patients. On the other hand, until recently, no solid evidence from large-scale randomized-controlled studies had been established the survival benefit of neoadjuvant (preoperative) therapy. In 2018 to 2019, one phase III study each of neoadjuvant therapy was conducted in Japan and overseas (Table 1).

The results of the phase III study (Prep-02/JSAP-05 Study) of neoadjuvant chemotherapy with gemcitabine plus S-1 for pancreatic cancer patients scheduled for resection conducted in Japan were reported at the American Society of Clinical Oncology-—Gastrointestinal Cancers Symposium (ASCO-GI) 2019; the study showed that the overall survival (OS) was significantly better in the neoadjuvant therapy group as compared to that in the upfront surgery group [hazard ratio (HR) 0.72, p = 0.015] [3‐5]. Approximately 80% of patients enrolled in this study had resectable pancreatic cancer at diagnosis, although the study also included patients with borderline resectable pancreatic cancer with portal vein invasion. A subgroup analysis showed better treatment outcomes in patients with resectable tumor and a trend towards better survival in patients with borderline resectable cancer assigned to the neoadjuvant therapy arm [5].

The overseas phase III study (PREOPANC-1 study) to confirm the survival benefit of neoadjuvant chemoradiotherapy was conducted in The Netherlands in patients with resectable and borderline resectable pancreatic cancer [6]. The results, reported at ASCO 2018, showed a trend towards a better OS in the neoadjuvant therapy arm as compared to the immediate surgery arm, although the difference was not statistically significant (HR 0.74, p = 0.074).

Prior to these two studies mentioned above, no phase III study had demonstrated the benefits of neoadjuvant therapy for patients with resectable pancreatic cancer [7, 8]; therefore, Japanese guidelines had not recommended neoadjuvant therapy for patients with pancreatic cancer until recently [9, 10]; the same is true of guidelines in other countries overseas, which still do not recommend neoadjuvant therapy as standard treatment for patients with resectable pancreatic cancer with exceptions for those with high-risk factors [11‐14]. Based on the results of the Prep-02/JSAP-05 study, however, the latest Japanese guidelines (Clinical Practice Guidelines for Pancreatic Cancer 2019) recommend gemcitabine plus S-1 combination therapy (GS therapy) as a standard neoadjuvant therapy for patients with resectable pancreatic cancer [15, 16]. Since the Prep-02/JSAP-05 study was conducted only in Japan and use of S-1 is not as feasible in Western populations, as mentioned later, until now, GS therapy is recognized as a standard therapy only in Japan and China [13, 17, 18].

For patients with borderline resectable pancreatic cancer, Japanese guidelines recommend neoadjuvant therapy, in general, but have refrained from recommending any specific regimens [15, 16]. Although guidelines in many other countries also recommend neoadjuvant therapy for borderline resectable pancreatic cancer, no consensus on any standard regimens has been established in any country until date [11‐14]. Among the several ongoing randomized-controlled trials of treatments for borderline resectable pancreatic cancer (Table 2) [19‐27], a phase II/III study of neoadjuvant therapy with gemcitabine plus nab-paclitaxel therapy versus chemoradiotherapy with S-1 is under way in Japan; this study is expected to provide specific new evidence for the establishment of a standard regimen for borderline resectable pancreatic cancer [21].

Randomized-controlled trials comparing postoperative adjuvant chemotherapy and resection alone have been conducted since the 1990s, mainly in Europe and Japan (Table 3). In the CONKO-001 trial conducted in Germany and Austria, 354 patients who had undergone resection for pancreatic cancer were randomly assigned to receive postoperative adjuvant chemotherapy with gemcitabine alone or resection alone [28, 29]. The results showed a significantly prolonged recurrence-free survival in the adjuvant chemotherapy arm. While no significant prolongation of the OS was noted initially (p = 0.06) [28], a subsequent analysis performed after long-term follow-up revealed significant prolongation of not only the recurrence-free survival, but also of the OS [29]. In the JSAP-02 study conducted in Japan, 118 patients who had undergone resection for pancreatic cancer were randomly assigned to receive postoperative adjuvant chemotherapy with gemcitabine alone or resection alone [30]. Consistent with the initial results of the CONKO-001 trial, significant prolongation of the recurrence-free survival was observed in the gemcitabine-alone arm. The European Study Group of Pancreatic Cancer (ESPAC) conducted the ESPAC-3 Study in Europe, Australia, Japan, and Canada, in which 1088 patients who had undergone resection for pancreatic cancer were randomly assigned to receive postoperative adjuvant chemotherapy with either fluorouracil plus folinate calcium or gemcitabine alone [31]. Although there was no significant difference in the OS between the two groups, the incidence of serious adverse events was significantly lower in the gemcitabine-alone arm than in the fluorouracil plus folinate calcium arm. These results indicate that patients receiving postoperative adjuvant chemotherapy with gemcitabine show significantly better survival outcomes than those undergoing resection alone; in addition, since serious adverse events were also less frequent in the gemcitabine arm than in the fluorouracil plus folinate calcium arm, gemcitabine could be regarded as the global standard treatment agent for postoperative adjuvant chemotherapy.

In Japan, the Japan Adjuvant Study Group of Pancreatic Center (JASPAC) conducted a phase III comparative study (JASPAC 01) of postoperative adjuvant chemotherapy with gemcitabine alone versus S-1 alone in patients who had undergone resection for pancreatic cancer [32]. A total of 385 patients were enrolled, and the 5-year survival rate and median survival time were 44.1% and 46.5 months, respectively, in the S-1 group, and 24.4% and 25.5 months, respectively, in the gemcitabine group. The results demonstrated that postoperative adjuvant therapy with S-1 as compared to that with gemcitabine was associated with a significantly improved OS after resection of pancreatic cancer (HR 0.57, p < 0.0001). The ESPAC conducted the ESPAC-4 study, in which 730 patients who had undergone resection for pancreatic cancer were randomly assigned to receive either gemcitabine alone or combined gemcitabine plus capecitabine therapy in England, Scotland, Wales, Germany, France, and Sweden [33]. The median survival time was 25.5 months in the gemcitabine monotherapy arm and 28.0 months in the gemcitabine plus capecitabine arm. The results demonstrated that the combined gemcitabine plus capecitabine regimen yielded a significantly prolonged OS after pancreatic cancer resection as compared to gemcitabine monotherapy (HR: 0.82, p = 0.032). The results of the PRODIGE 24-ACCORD 24/CCTG PA 6 study, conducted in France and Canada, have also been reported; in this study, the modified FOLFIRINOX regimen was compared with gemcitabine alone as adjuvant therapy [34]. A total of 493 patients were enrolled, and the median disease-free survival, which was the primary endpoint, was 21.6 months in the modified FOLFIRINOX arm and 12.8 months in the gemcitabine monotherapy arm, demonstrating superior outcomes in the modified FOLFIRINOX arm (HR 0.58, p < 0.0001). In terms of the OS also, better results were obtained in the modified FOLFIRINOX arm (the median survival time was 54.4 months in the modified FOLFIRINOX arm and 35.0 months in the gemcitabine monotherapy arm; HR 0.64, p = 0.003). Since no clinical study has been conducted to compare S-1 alone with the combined gemcitabine plus capecitabine regimen and modified FOLFIRINOX regimen, and it is still not clear as to which of the three above regimes might be the optimal one for adjuvant therapy. In Japan and China, S-1 is frequently used as the standard treatment agent [13, 15, 16] because of its higher efficacy as compared to gemcitabine monotherapy (as suggested by the superior HR of 0.57) [32], its milder adverse effects in Asians, and its availability as an oral formulation, which can be expected to reduce the burden on the patients. On the other hand, S-1 has not been tested as adjuvant therapy in Western populations [12, 17]. Therefore, in countries including Europe and the US, the gemcitabine plus capecitabine regimen or modified FOLFIRINOX regimen is preferred and regarded as the standard [11, 12, 14]. Thus, this is another difference in the treatment practice for pancreatic cancer between Asian and Western countries.

The global phase III APACT trial evaluated adjuvant treatment with nab-paclitaxel plus gemcitabine versus gemcitabine alone in patients with resected pancreatic cancer [35]. Results of the study were reported at ASCO 2019. The primary endpoint, disease-free survival by independent review, was not met. The median disease-free survival was 19.4 months with nab-paclitaxel plus gemcitabine versus 18.8 months with gemcitabine monotherapy (HR = 0.88; p = 0.1824). However, the prespecified sensitivity analysis of investigator-assessed disease-free survival and interim OS were improved with nab-paclitaxel plus gemcitabine versus gemcitabine alone (HR 0.82 for both). Additional OS follow-up may better support nab-paclitaxel plus gemcitabine as an option in the adjuvant setting.

The Japanese Clinical Practice Guidelines for Pancreatic Cancer 2019 recommends FOLFIRINOX therapy or combined gemcitabine plus nab-paclitaxel therapy as the first-line treatment for pancreatic cancer patients with distant metastases [15, 16]. For patients in whom these treatments are unsuitable owing to their systemic condition or age, gemcitabine monotherapy, S-1 monotherapy or gemcitabine plus erlotinib combination therapy is recommended. A phase III study conducted overseas demonstrated the survival benefits of the FOLFIRINOX regimen [37], combined gemcitabine plus nab-paclitaxel regimen [38], gemcitabine monotherapy [40], and the gemcitabine plus erlotinib regimen [41]. Thereafter, clinical trials were also conducted in Japan, and the efficacy and safety of these regimens were also confirmed in Japanese patients [42‐45]. On the other hand, S-1 monotherapy has come to be recommended as a standard treatment on the basis of the results of a phase III study (GEST study) conducted in Japan and Taiwan [46, 47]. The GEST study examined the non-inferiority of S-1 monotherapy to gemcitabine monotherapy and the superiority of combined gemcitabine plus S-1 therapy over gemcitabine monotherapy, in terms of the survival outcomes. The non-inferiority of S-1 monotherapy was statistically confirmed (the median survival time was 8.8 months in the gemcitabine arm and 9.7 months in the S-1 arm; HR 0.96, p < 0.001). However, the superiority of the combined gemcitabine plus S-1 regimen could not be confirmed (the median survival time was 8.8 months in the gemcitabine arm and 10.1 months in the gemcitabine plus S-1 arm; HR 0.88, p = 0.15). Therefore, S-1 monotherapy has come to be regarded as a standard treatment in Asian countries [13, 15, 16]. However, S-1 therapy has been reported to have only marginal anti-tumor activity in Western populations [48], and is, therefore, not regarded as a valid treatment option in Western countries.

The second-line treatment regimens recommended in Japan are: (1) a fluorouracil-containing regimen after a first-line gemcitabine-containing regimen, (2) a gemcitabine-containing regimen after a first-line fluorouracil-containing regimen, and (3) pembrolizumab for microsatellite instability-high (MSI-H) [15, 16]. The fluorouracil-containing regimen consists of fluorouracil combined with folinate calcium and MM-398 (nanoliposomal irinotecan).

Tropomyosin receptor kinase (TRK) inhibitors have been reported to be useful for patients with solid tumors harboring neurotrophic receptor tyrosine kinase (NTRK) fusion genes. At the 2018 Annual Meeting of the European Society of Clinical Oncology (ESMO), the results from an integrated analysis of three studies (STARTRK-2 study, STARTRK-1 study and ALKA-372-001 study) conducted in patients with tumors harboring NTRK fusion genes were presented [49]. In these studies, entrectinib, a TRK inhibitor, was administered to 54 patients with soft-tissue sarcomas, non-small cell lung cancer, and others, and the response rate was 57.4%. Entrectinib was also approved in Japan in June 2019. Pancreatic cancer harboring NTRK fusion genes is reported to be extremely rare, accounting for less than 1% of all cases [50]. The Japanese guidelines are being revised to include entrectinib as a treatment option.

KRAS, CDKN2A, TP53, and SMAD4 have been recognized as major driver genes in pancreatic carcinogenesis. Mutations in these genes are the most commonly encountered mutations in the majority of patients with pancreatic cancer, although potential therapeutic-target genes are limited to KRAS G12C and CDKN2A, which are found only in a small subgroup of patients. Thus, at present, there are no promising therapeutic agents for non-KRAS G12C, TP53, and SMAD4-mutated pancreatic cancer, which account for the majority of the cases. Therefore, selection of the treatment regimen based on gene mutations has yet to become standard strategies for patients with pancreatic cancer. Of the gene mutations, mutations of the DNA repair genes, such as BRCA1/2, PALB2, ATM, ATR, and ATRX, are the most common “highly actionable” alterations. It has been reported that in patients with gene mutations leading to homologous recombination deficiency (HRD), poly (ADP-ribose) polymerase (PARP) inhibitors exert anti-tumor effects by inducing cell death via synthetic lethality. In recent years, promising results have been reported from clinical studies of PARP inhibitors for patients with germline BRCA-mutated pancreatic cancer. According to the findings of the recently completed, international, phase-III POLO (Pancreas cancer OLaparib Ongoing) trial, treatment with the PARP inhibitor olaparib significantly reduced the risk of disease progression in patients with a germline BRCA1 or BRCA2 mutation and metastatic pancreatic cancer and disease that had not progressed during the first-line platinum-based chemotherapy [78]. In this study, 3315 patients with pancreatic cancer were screened for germline BRCA1/2 mutations, and 247 (7.5%) were found to have BRCA1/2 mutations. Of these 247 patients, 154 were randomized to receive olaparib or placebo. The primary endpoint, namely, progression-free survival, was significantly prolonged in the olaparib group as compared to the placebo group (median progression-free survival: 7.4 months vs. 3.8 months, HR 0.53, p = 0.004). Based on this result, the National Comprehensive Cancer Network (NCCN) guidelines in the United States includes the recommendation of olaparib as maintenance treatment for patients who have germline BRCA1/2 mutations, good PS, metastatic disease, and no disease progression after at least 4–6 month first-line chemotherapy [11]. Other PARP inhibitors such as veliparib and rucaparib are also being examined in clinical trials for pancreatic cancer [79‐81].

Although KRAS mutations are predominant in pancreatic cancer, no effective therapeutic agent targeting KRAS mutations has been discovered until date. Recently, however, several candidate inhibitors of the KRAS G12C mutant protein have been reported. Among these, encouraging results were reported from a phase I study of AMG510, especially in a non-small cell lung cancer cohort [82]. On the other hand, pancreatic cancers with wild-type KRAS, although rare, account for only about 5% of all cases; BRAF and EGFR gene mutations and some fusion genes (FGFR, ALK, NTRK, NRG1, etc.) have also been reported to be detected at a relatively high frequency [83].

The “Know Your Tumor Project” initiative undertaken by the Pancreatic Cancer Patient Association (PanCAN) in the United States reported that 26% of patients with pancreatic cancer had actionable alterations, suggesting the possibility that therapy targeting these alterations could prolong the survival [84]. Recently, the NCCN guidelines have been updated to include the following: “Germline testing is recommended for any patient with confirmed pancreatic cancer using comprehensive gene panels for hereditary cancer syndromes” and “Tumor/somatic gene profiling is recommended for patients with locally advanced/metastatic disease who are candidates for anti-cancer therapy to identify uncommon mutations.” Thus, the guidelines recommend a search for actionable mutations, even if they are rare [11]. In the near future, guidelines in other countries may also follow these recommendations.

Metastatic pancreatic ductal adenocarcinoma is characterized by excessive hyaluronan (HA) accumulation in the tumor microenvironment, with elevation of the interstitial pressure and impaired perfusion. Preclinical studies have demonstrated that pegvorhyaluronidase alfa (PEGPH20) degrades HA, thereby increasing drug delivery. In a randomized phase II study, patients with previously untreated metastatic pancreatic ductal adenocarcinoma were randomly assigned to treatment with PEGPH20 plus nab-paclitaxel/gemcitabine (PAG) or nab-paclitaxel/gemcitabine (AG) [85]. The progression-free survival was significantly prolonged with PAG treatment in the overall subject population (HR 0.73; p = 0.049) and in patients with HA-high tumors (HR 0.51; p = 0.048). On the other hand, PEGPH20 combination with modified FOLFIRINOX caused increased toxicity and resulted in decreased treatment duration compared with modified FOLFIRINOX alone in another randomized study [86]. A global phase III study of PAG versus AG in patients with HA-high PDA is now ongoing [87].

Circulating tumor cells (CTCs) and circulating tumor DNA (ctDNA), as liquid biopsies, are an emerging minimally invasive tool with a unique potential for (a) determining the prognosis, (b) monitoring therapeutic responses and tumor recurrence in real time, (c) exploring therapeutic targets, and (d) potentially developing new drugs by studying metastatic cancer biology and drug resistance mechanisms [88‐90]. In addition to CTCs and ctDNA, circulating tumor extracellular vesicles (e.g., exosomes), tumor-educated platelets (TEPs), and blood-based protein and metabolite markers also show early promise as biomarkers that could be used from cancer screening through to targeted treatments for pancreatic cancer. In particular, ctDNA is expected to be clinically useful in noninvasive molecular profiling for the novel actionable mutations [91]. Sequential real-time liquid biopsies could potentially allow early identification of resistance to cancer therapy in individual patients as an important hallmark of personalized cancer medicine. Detection and characterization of minimal residual disease after resection are another important aim and challenge for future studies [92].

Steady progress has been made in the development of non-surgical therapies for advanced pancreatic cancer, and the prognosis of the patients is steadily improving, although the results are still far from satisfactory. Until now, the standard systemic treatment for pancreatic cancer has been limited to existing cytotoxic anti-cancer drugs. Recently, large-scale randomized-controlled studies of pre- and postoperative adjuvant therapies have been actively undertaken in an attempt to improve the prognosis of patients with resectable pancreatic cancer by introducing chemotherapies which have been commonly used for advanced pancreatic cancer, with or without radiotherapy, and this trend is expected to continue in the future. For some pancreatic cancers with actionable mutations, such as germline BRCA1/2 mutations, NTRK fusion mutation, and MSI-H, molecular-targeted therapy and immunotherapy are being introduced. It is expected that the mechanisms of initiation and development of pancreatic cancer will be better elucidated and that targeted treatments that would yield marked tumor shrinkage and survival prolongation will be developed. In addition to developing highly effective and safe anti-cancer treatments, it is also important to ensure that supportive and palliative care is available for efficient implementation of such anti-cancer treatments, and multidisciplinary cooperation and collaboration with the community should be promoted, because both improvement of the patient prognosis and improvement of the quality of life are major goals of treatment of this disease.